Update implementation of Basic02 #1197

tgreenx · 2023-02-22T10:11:48Z

Purpose

This PR proposes an update implementation of Basic02 to latest specification (zonemaster/zonemaster#1085 and zonemaster/zonemaster#1145).

Context

Fixes #1193, #881

Changes

Update implementation of Basic02:

Logic
Message tags

Update unitary tests:

Some are entirely new
Modernized utility functions

How to test this PR

Unitary tests should pass.

Also (without an IPv6 connectivity):

$ zonemaster-cli --test basic/basic02 zonemaster.net --level INFO --show-testcase --raw --no-ipv6
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.2
   6.09 INFO      BASIC02        B02_AUTH_RESPONSE_SOA   domain="zonemaster.net"; ns_list=ns2.nic.fr/192.93.0.4;nsa.dnsnode.net/194.58.192.46;nsp.dnsnode.net/194.58.198.32;nsu.dnsnode.net/185.42.137.98

And (from #881):

$ zonemaster-cli --test basic/basic02 e.consistency05.exempelvis.se --level INFO --show-testcase --raw --no-ipv6
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.2
   1.37 CRITICAL  BASIC02        B02_NO_WORKING_NS   domain="e.consistency05.exempelvis.se"
   1.37 ERROR     BASIC02        B02_UNEXPECTED_RCODE   ns=ns2.e.consistency05.exempelvis.se/194.18.226.122; rcode=NXDOMAIN
   1.37 ERROR     BASIC02        B02_UNEXPECTED_RCODE   ns=ns1.e.consistency05.exempelvis.se/46.21.97.97; rcode=REFUSED

matsduf · 2023-02-23T15:06:42Z

t/util.t .................. 1/? unable to parse RR string at /usr/local/lib/perl5/site_perl/Net/DNS/ZoneFile.pm line 534.
 at /usr/home/matsd/.cpanm/work/1677164454.51730/Zonemaster-Engine-v4.6.1/blib/lib/Zonemaster/Engine/Util.pm line 179.

tgreenx · 2023-02-23T15:15:16Z

t/util.t .................. 1/? unable to parse RR string at /usr/local/lib/perl5/site_perl/Net/DNS/ZoneFile.pm line 534.
 at /usr/home/matsd/.cpanm/work/1677164454.51730/Zonemaster-Engine-v4.6.1/blib/lib/Zonemaster/Engine/Util.pm line 179.

This error is also present on master and develop branches, actually.

matsduf · 2023-02-23T15:54:09Z

I tested the name lame.dufberg.se that is a lame delegation. The expected message is B02_NO_WORKING_NS and B02_NS_NO_RESPONSE but that is not what we get.

# zonemaster-cli --test basic/basic02 lame.dufberg.se --level INFO --show-testcase --raw
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.1
  20.86 CRITICAL  BASIC02        B02_NO_DELEGATION   domain="lame.dufberg.se"

matsduf · 2023-02-23T15:57:54Z

This error is also present on master and develop branches, actually.

You are right. That is a false error message. It should be hidden.

matsduf · 2023-02-23T16:03:24Z

lib/Zonemaster/Engine/Test/Basic.pm

+        # This is not a realistical conditional check considering the current implementation of Engine::Nameserver.
+        # Any Engine::Nameserver object created will (must) have an IP address. So it is here as a placeholder.
+        if ( not $ns->address ) {
+            $ns_cant_resolve{$ns->name} = 1;
+            next;
+        }


That limitation is really bad. That means that we cannot see when a domain is delegated to a name server name that cannot be resolved into an address.

Yes, that is my understanding as well. I will see what I can do about it.

matsduf · 2023-02-23T16:06:35Z

lib/Zonemaster/Engine/Test/Basic.pm

+    my %ns_cant_resolve;
+    my %ns_no_response;
+    my %unexpected_rcode;
+
    my @ns = @{ Zonemaster::Engine::TestMethods->method4( $zone ) };


Does metod4 return an empty array when the delegation is completely lame?

Yes, that is currently the case.

$ git diff diff --git a/lib/Zonemaster/Engine/Test/Basic.pm b/lib/Zonemaster/Engine/Test/Basic.pm index 714b7176..18e45f25 100644 --- a/lib/Zonemaster/Engine/Test/Basic.pm +++ b/lib/Zonemaster/Engine/Test/Basic.pm @@ -395,6 +395,18 @@ sub basic02 { my %ns_no_response; my %unexpected_rcode; + say $zone->name; + say $zone->parent->name; + say "glue names"; + say @{$zone->glue_names}; + say "glue"; + say @{$zone->glue}; + say "ns names"; + say @{$zone->ns_names}; + say "ns"; + say @{$zone->ns}; + die; + my @ns = @{ Zonemaster::Engine::TestMethods->method4( $zone ) }; if ( not scalar @ns ) { $ zonemaster-cli --show-testcase --test basic/basic02 lame.dufberg.se --level INFO --no-ipv6 --raw 0.00 INFO UNSPECIFIED GLOBAL_VERSION version=v4.6.1 lame.dufberg.se dufberg.se glue names ns1.lame.dufberg.sens2.lame.dufberg.se glue ns names ns 21.25 CRITICAL UNSPECIFIED MODULE_ERROR module=Basic; msg=Died at /home/tgreen/zonemaster/zonemaster-engine/lib/Zonemaster/Engine/Test/Basic.pm line 408. $ zonemaster-cli --show-testcase --test basic/basic02 zonemaster.fr --level INFO --no-ipv6 --raw 0.00 INFO UNSPECIFIED GLOBAL_VERSION version=v4.6.1 zonemaster.fr fr glue names ns1.nic.frns3.nic.fr glue ns1.nic.fr/192.134.4.1ns1.nic.fr/2001:67c:2218:2::4:1ns3.nic.fr/192.134.0.49ns3.nic.fr/2001:660:3006:1::1:1 ns names ns1.nic.frns3.nic.fr ns ns1.nic.fr/192.134.4.1ns1.nic.fr/2001:67c:2218:2::4:1ns3.nic.fr/192.134.0.49ns3.nic.fr/2001:660:3006:1::1:1 0.51 CRITICAL UNSPECIFIED MODULE_ERROR module=Basic; msg=Died at /home/tgreen/zonemaster/zonemaster-engine/lib/Zonemaster/Engine/Test/Basic.pm line 408.

So "glue_names" (aka. Method2) is non-empty and has the right name server names for the zone. However, "glue" (i.e, the creation of name server objects, aka. Method4) is empty. The function that populates the "glue" is a special array object Engine::NSArray, filled using Engine::NSArray::_load_name(), which itself calls Engine::Recursor::get_addresses_for(), and in turn, Engine::Recursor::_recurse() which skips unresponsive name servers.

I'll investigate further though.

Yes, that is currently the case.

That seems to be the wrong method.

So "glue_names" (aka. Method2) is non-empty and has the right name server names for the zone. However, "glue" (i.e, the creation of name server objects, aka. Method4) is empty.

So why not use Method2 to get the delegation (NS) and Method4 to get the glue address records (does the implemented Method4 include non-glue address records)? The specification says:

1. Retrieve the NS records for *Child Zone* using [Method 2] and the IP addresses ([glue records][glue record]) for any [in-bailiwick] name servers using [Method 4]. 2. Retrieve the IP addresses for any [out-of-bailiwick] name servers using recursive lookup for address records (both IPv4 and IPv6) for that name and add resolved addresses, if any, to the set.

Mats

does the implemented Method4 include non-glue address records

Digging into the implementation, I would answer yes (however, I've not tested it yet):

zonemaster-engine/lib/Zonemaster/Engine/TestMethods.pm

Lines 40 to 44 in 0fdd780

sub method4 {

my ( $class, $zone ) = @_;

return $zone->glue;

}

zonemaster-engine/lib/Zonemaster/Engine/Zone.pm

Lines 56 to 77 in 0fdd780

sub _build_glue {

my ( $self ) = @_;

my @glue_names = @{ $self->glue_names };

my $zname = $self->name->string;

if ( Zonemaster::Engine::Recursor->has_fake_addresses( $zname ) ) {

my @ns_list;

foreach my $ns ( @glue_names ) {

foreach my $ip ( Zonemaster::Engine::Recursor->get_fake_addresses( $zname, $ns ) ) {

push @ns_list, Zonemaster::Engine::Nameserver->new( { name => $ns, address => $ip } );

}

}

return \@ns_list;

}

else {

my $aref = [];

tie @$aref, 'Zonemaster::Engine::NSArray', @glue_names;

return $aref;

}

}

zonemaster-engine/lib/Zonemaster/Engine/NSArray.pm

Lines 137 to 152 in 0fdd780

=head1 NAME

Zonemaster::Engine::NSArray - Class implementing arrays that lazily looks up name server addresses from their names

=head1 SYNOPSIS

tie @ary, 'Zonemaster::Engine::NSArray', @ns_names

=head1 DESCRIPTION

This class is used for the C<glue> and C<ns> attributes of the

L<Zonemaster::Engine::Zone> class. It is initially seeded with a list of

names, which will be expanded into proper L<Zonemaster::Engine::Nameserver>

objects on demand. Be careful with using Perl functions that act on

whole arrays (particularly C<foreach>), since they will usually force

the entire array to expand, negating the use of the lazy-loading.

ref: https://perldoc.perl.org/perltie (I'm discovering this tie function, never used nor tested it yet.)

does the implemented Method4 include non-glue address records

Digging into the implementation, I would answer yes (however, I've not tested it yet):

Well, then it is probably enough if the implementation uses Method2 to get the NS in the delegation and Method4 to get nameserver IP addresses using Method4.

tgreenx · 2023-03-07T16:20:23Z

@matsduf @PNAX I think I got to a suitable workaround for both issues. See commit 9e700c7. Let me know what you think (note that unitary tests data are still not updated and thus will fail).

$ zonemaster-cli --show-testcase --test basic/basic02 lame.dufberg.se --level INFO --no-ipv6 --raw
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.1
  20.94 CRITICAL  BASIC02        B02_NO_WORKING_NS   domain="lame.dufberg.se"
  20.94 WARNING   BASIC02        B02_NS_NO_RESPONSE   ns=ns2.lame.dufberg.se/188.126.83.221
  20.94 WARNING   BASIC02        B02_NS_NO_RESPONSE   ns=ns1.lame.dufberg.se/159.253.30.207

matsduf · 2023-03-08T12:27:55Z

@matsduf @PNAX I think I got to a suitable workaround for both issues. See commit 9e700c7. Let me know what you think (note that unitary tests data are still not updated and thus will fail).

I am not sure that it is a work-around or just the solution for the methods that are available, but it looks fine. I will create some more test delegations for related scenarios to see if they behave the same.

matsduf · 2023-03-08T14:12:50Z

I tested the following test zones and I can see that the code behaves correctly. Unfortunately, the specification calls for checking for the AA flag before checking for the RCODE, but I think we can live with that for now.

$ zonemaster-cli --show-testcase --test basic/basic02 lame-ns-no-name.dufberg.se. --level INFO --raw
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.1
   0.45 CRITICAL  BASIC02        B02_NO_WORKING_NS   domain="lame-ns-no-name.dufberg.se"
   0.45 ERROR     BASIC02        B02_NS_NO_IP_ADDR   nsname=ns1.test.xa

$ zonemaster-cli --show-testcase --test basic/basic02 lame-ns-no-addr.dufberg.se --level INFO --raw
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.1
   0.49 CRITICAL  BASIC02        B02_NO_WORKING_NS   domain="lame-ns-no-addr.dufberg.se"
   0.49 ERROR     BASIC02        B02_NS_NO_IP_ADDR   nsname=ns.se

$ zonemaster-cli --show-testcase --test basic/basic02 lame-ns-no-response.dufberg.se --level INFO --raw
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.1
  10.55 CRITICAL  BASIC02        B02_NO_WORKING_NS   domain="lame-ns-no-response.dufberg.se"
  10.55 WARNING   BASIC02        B02_NS_NO_RESPONSE   ns=ns1.lame-ns-no-response.dufberg.se/159.253.30.207

$ zonemaster-cli --show-testcase --test basic/basic02 lame-ns-refused.dufberg.se. --level INFO --raw
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.1
   0.51 CRITICAL  BASIC02        B02_NO_WORKING_NS   domain="lame-ns-refused.dufberg.se"
   0.51 ERROR     BASIC02        B02_NS_NOT_AUTH   ns=nsa.dnsnode.net/2a01:3f1:46::53
   0.51 ERROR     BASIC02        B02_NS_NOT_AUTH   ns=nsa.dnsnode.net/194.58.192.46

tgreenx · 2023-03-08T15:42:11Z

I tested the following test zones and I can see that the code behaves correctly. Unfortunately, the specification calls for checking for the AA flag before checking for the RCODE, but I think we can live with that for now.

[...]

Good to hear! If you want, I can make a quick update to the specification and swap those two checks around, so that the implementation can be updated too. It seems like a very small thing to do.

Also, do you want me to add those zones to the unitary tests?

matsduf · 2023-03-08T20:17:35Z

I created yet another test zone (or rather test delegation):

zonemaster-cli --show-testcase --test basic/basic02 lame-ns-servfail.dufberg.se. --level INFO --raw
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.1
   0.55 CRITICAL  BASIC02        B02_NO_WORKING_NS   domain="lame-ns-servfail.dufberg.se"
   0.55 ERROR     BASIC02        B02_NS_NOT_AUTH   ns=dns3.narnia.pp.se/2a05:d014:4ae:e900:e026:2d09:8d68:2d43
   0.55 ERROR     BASIC02        B02_NS_NOT_AUTH   ns=dns3.narnia.pp.se/3.124.111.178

Good to hear! If you want, I can make a quick update to the specification and swap those two checks around, so that the implementation can be updated too. It seems like a very small thing to do.

Please review zonemaster/zonemaster#1145

Also, do you want me to add those zones to the unitary tests?

Yes, that would be good.

After an update based on zonemaster/zonemaster#1145, the following zones should return B02_NS_NO_IP_ADDR:

lame-ns-no-name.dufberg.se.
lame-ns-no-addr.dufberg.se.
lame-ns-no-glue.dufberg.se.

The following B02_NS_NO_RESPONSE:

lame-ns-no-response.dufberg.se.

The following B02_UNEXPECTED_RCODE:

lame-ns-refused.dufberg.se.
lame-ns-servfail.dufberg.se.

Update implementation of Basic02 to latest specification (Zonemaster PR#1085) Update unitary tests: - some are entirely new, so unitary tests data needs an update but it has not been done yet - modernized utility functions

- Handling of lame delegations by making use of Method2 (i.e. Engine::Zone::_build_glue_names()) and Engine::Zone::_build_glue_addresses(). It is needed because Engine::Recursor::_recurse(), used by Engine::Recursor::get_addresses_for(), ignores unresponsive name servers when doing recursive lookups, which then leads to empty Engine::Zone attributes ('glue', and thus, 'ns_names' and 'ns') when using Method4. - Handling of name servers that cannot be resolved in an IP address.

- Update to latest specification (Zonemaster PR#1145) - Add unitary tests

tgreenx · 2023-03-09T10:08:30Z

[ ... ]

Please review zonemaster/zonemaster#1145

Done, and implementation is updated accordingly.

Also, do you want me to add those zones to the unitary tests?

Yes, that would be good.

[ ... ]

Done. Also rebased on latest develop.

The only remaining thing is to update unitary test data. I will do that soon.

tgreenx · 2023-03-09T15:08:22Z

Unitary tests data are updated, and checks are now passing. Please re-review.

It looks good now.

matsduf · 2023-03-09T15:48:34Z

Here is another test (expecting B02_NS_BROKEN):

zonemaster-cli --show-testcase --test basic/basic02  --level INFO --raw www.zonemaster.net --ns nsa.dnsnode.net
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.2
   0.11 CRITICAL  BASIC02        B02_NO_WORKING_NS   domain="www.zonemaster.net"
   0.11 ERROR     BASIC02        B02_NS_BROKEN   ns=nsa.dnsnode.net/2a01:3f1:46::53
   0.11 ERROR     BASIC02        B02_NS_BROKEN   ns=nsa.dnsnode.net/194.58.192.46

tgreenx · 2023-03-09T17:27:56Z

Here is another test:

[ ... ]

I added it to the unitary tests and updated unitary test data.

matsduf · 2023-03-09T22:00:03Z

This delegation expects B02_NS_NOT_AUTH:

$ zonemaster-cli --show-testcase --test basic/basic02  --level INFO --raw zonemaster.net --ns g.root-servers.net --ns a.root-servers.net
   0.00 INFO      UNSPECIFIED    GLOBAL_VERSION   version=v4.6.2
   0.26 CRITICAL  BASIC02        B02_NO_WORKING_NS   domain="zonemaster.net"
   0.26 ERROR     BASIC02        B02_NS_NOT_AUTH   ns=g.root-servers.net/2001:500:12::d0d
   0.26 ERROR     BASIC02        B02_NS_NOT_AUTH   ns=a.root-servers.net/2001:503:ba3e::2:30
   0.26 ERROR     BASIC02        B02_NS_NOT_AUTH   ns=a.root-servers.net/198.41.0.4
   0.26 ERROR     BASIC02        B02_NS_NOT_AUTH   ns=g.root-servers.net/192.112.36.4

tgreenx · 2023-03-14T10:34:37Z

This delegation expects B02_NS_NOT_AUTH:

[ ... ]

Thanks, I included this test in the unitary tests (commit b17fca5). Now we have full testing coverage for Basic02.
Please re-review.

matsduf · 2023-03-14T10:52:47Z

Now we have full testing coverage for Basic02. Please re-review.

That is really great.

tgreenx added A-TestCase Area: Test case specification or implementation of test case V-Minor Versioning: The change gives an update of minor in version. labels Feb 22, 2023

tgreenx added this to the v2023.1 milestone Feb 22, 2023

tgreenx requested review from matsduf, vlevigneron, hannaeko, a user and marc-vanderwal February 22, 2023 10:11

tgreenx linked an issue Feb 22, 2023 that may be closed by this pull request

Basic02 gives incorrect error message when zone is completely lame #881

Closed

matsduf previously requested changes Feb 23, 2023

View reviewed changes

tgreenx linked an issue Mar 7, 2023 that may be closed by this pull request

Update implementation of updated BASIC02 #1193

Closed

tgreenx requested a review from matsduf March 7, 2023 16:20

tgreenx added 3 commits March 9, 2023 10:50

Initial commit - Update implementation of Basic02

82977a7

Update implementation of Basic02 to latest specification (Zonemaster PR#1085) Update unitary tests: - some are entirely new, so unitary tests data needs an update but it has not been done yet - modernized utility functions

Update implementation of Basic02

d6e979c

- Update to latest specification (Zonemaster PR#1145) - Add unitary tests

tgreenx force-pushed the update-basic02 branch from 9e700c7 to d6e979c Compare March 9, 2023 09:52

tgreenx force-pushed the update-basic02 branch from e2a7739 to 8adff0e Compare March 9, 2023 17:16

tgreenx force-pushed the update-basic02 branch from 8adff0e to 6d5ba8a Compare March 9, 2023 17:22

matsduf previously approved these changes Mar 9, 2023

View reviewed changes

tgreenx dismissed matsduf’s stale review via 9690b89 March 14, 2023 10:32

tgreenx force-pushed the update-basic02 branch from 6d5ba8a to 9690b89 Compare March 14, 2023 10:32

tgreenx requested a review from matsduf March 14, 2023 10:32

Update unitary tests and data for Basic02

b17fca5

matsduf previously approved these changes Mar 14, 2023

View reviewed changes

tgreenx dismissed matsduf’s stale review via b17fca5 March 14, 2023 10:52

tgreenx force-pushed the update-basic02 branch from 9690b89 to b17fca5 Compare March 14, 2023 10:52

matsduf approved these changes Mar 14, 2023

View reviewed changes

tgreenx merged commit 6bf8c71 into zonemaster:develop Apr 5, 2023

tgreenx deleted the update-basic02 branch April 5, 2023 14:10

This was referenced Apr 19, 2023

Update implementation of updated BASIC02 #1193

Closed

Basic02 gives incorrect error message when zone is completely lame #881

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update implementation of Basic02 #1197

Update implementation of Basic02 #1197

tgreenx commented Feb 22, 2023 •

edited

Loading

matsduf commented Feb 23, 2023

tgreenx commented Feb 23, 2023

matsduf commented Feb 23, 2023

matsduf commented Feb 23, 2023

matsduf Feb 23, 2023

tgreenx Feb 23, 2023

matsduf Feb 23, 2023

tgreenx Feb 23, 2023

matsduf Feb 24, 2023

ghost Feb 28, 2023

matsduf Mar 6, 2023

tgreenx commented Mar 7, 2023 •

edited

Loading

matsduf commented Mar 8, 2023

matsduf commented Mar 8, 2023

tgreenx commented Mar 8, 2023

matsduf commented Mar 8, 2023

tgreenx commented Mar 9, 2023 •

edited

Loading

tgreenx commented Mar 9, 2023 •

edited

Loading

matsduf commented Mar 9, 2023 •

edited

Loading

tgreenx commented Mar 9, 2023

matsduf commented Mar 9, 2023

tgreenx commented Mar 14, 2023 •

edited

Loading

matsduf commented Mar 14, 2023

	sub _build_glue {
	my ( $self ) = @_;
	my @glue_names = @{ $self->glue_names };
	my $zname = $self->name->string;

	if ( Zonemaster::Engine::Recursor->has_fake_addresses( $zname ) ) {
	my @ns_list;
	foreach my $ns ( @glue_names ) {
	foreach my $ip ( Zonemaster::Engine::Recursor->get_fake_addresses( $zname, $ns ) ) {
	push @ns_list, Zonemaster::Engine::Nameserver->new( { name => $ns, address => $ip } );
	}
	}
	return \@ns_list;
	}
	else {

	my $aref = [];
	tie @$aref, 'Zonemaster::Engine::NSArray', @glue_names;

	return $aref;
	}
	}

	=head1 NAME

	Zonemaster::Engine::NSArray - Class implementing arrays that lazily looks up name server addresses from their names

	=head1 SYNOPSIS

	tie @ary, 'Zonemaster::Engine::NSArray', @ns_names

	=head1 DESCRIPTION

	This class is used for the C<glue> and C<ns> attributes of the
	L<Zonemaster::Engine::Zone> class. It is initially seeded with a list of
	names, which will be expanded into proper L<Zonemaster::Engine::Nameserver>
	objects on demand. Be careful with using Perl functions that act on
	whole arrays (particularly C<foreach>), since they will usually force
	the entire array to expand, negating the use of the lazy-loading.

Update implementation of Basic02 #1197

Update implementation of Basic02 #1197

Conversation

tgreenx commented Feb 22, 2023 • edited Loading

Purpose

Context

Changes

How to test this PR

matsduf commented Feb 23, 2023

tgreenx commented Feb 23, 2023

matsduf commented Feb 23, 2023

matsduf commented Feb 23, 2023

matsduf Feb 23, 2023

Choose a reason for hiding this comment

tgreenx Feb 23, 2023

Choose a reason for hiding this comment

matsduf Feb 23, 2023

Choose a reason for hiding this comment

tgreenx Feb 23, 2023

Choose a reason for hiding this comment

matsduf Feb 24, 2023

Choose a reason for hiding this comment

ghost Feb 28, 2023

Choose a reason for hiding this comment

matsduf Mar 6, 2023

Choose a reason for hiding this comment

tgreenx commented Mar 7, 2023 • edited Loading

matsduf commented Mar 8, 2023

matsduf commented Mar 8, 2023

tgreenx commented Mar 8, 2023

matsduf commented Mar 8, 2023

tgreenx commented Mar 9, 2023 • edited Loading

tgreenx commented Mar 9, 2023 • edited Loading

matsduf commented Mar 9, 2023 • edited Loading

tgreenx commented Mar 9, 2023

matsduf commented Mar 9, 2023

tgreenx commented Mar 14, 2023 • edited Loading

matsduf commented Mar 14, 2023

tgreenx commented Feb 22, 2023 •

edited

Loading

tgreenx commented Mar 7, 2023 •

edited

Loading

tgreenx commented Mar 9, 2023 •

edited

Loading

tgreenx commented Mar 9, 2023 •

edited

Loading

matsduf commented Mar 9, 2023 •

edited

Loading

tgreenx commented Mar 14, 2023 •

edited

Loading